110 C. E. Gordon — Early Stages in Paleozoic Corals. 



been on the order of appearance of the septa, so far as could 

 be inferred by a careful study of early stages. It is by no 

 means true, however, that the early stages of all fossil forms 

 will give us the clue to the phylogeny of the group. It fre- 

 quently happens that certain individuals represent a special- 

 ized condition and are, on that account, not suitable for work- 

 ing out their own life history, which could be interpreted only 

 by a careful comparative study of many individuals. This 

 specialized condition may result from an acceleration in devel- 

 opment of certain ancestral characters resulting in a prema- 

 ture appearance of certain stages which normally belong some- 

 what later in the animal's life, and which obscure the early 

 conditions by their premature development ; again, certain 

 characters are retained longer than usual and are prolonged to 

 the obliteration of later stages, for the appearance of which 

 the life of the animal is not long enough. In interpreting the 

 'structural characteristics of any species the possibility of 

 specialization must not be overlooked. Among forms some- 

 what removed in geological time from the ancestral stock, one 

 must look for types in which acceleration, or retardation, or 

 both, have been important factors in altering early ancestral 

 conditions, and have produced new species, or varieties, 

 according to one's choice of designation. 



It is not within the province of man to say where, or in 

 what parts, these modifications are to take place when only 

 natural conditions operate to produce them. They are to be 

 looked for anywhere, and it is reasonable perhaps to expect 

 them in parts which have taken on new physiological import- 

 ance, or have ceased to have such importance, as responses to 

 certain conditions. Instances are numerous which forcibly 

 illustrate the principles of acceleration and retardation. The 

 genera Cyathophyllum and Heliophyllum are distinguished by 

 the presence in the latter of carinae. But sections of certain 

 Heliophyllums {JTelioj)hylhmi halli) show the carinae appear- 

 ing very late in the life of the individual.* Up to the time of 

 their appearance they are not to be distinguished from Cyatho- 

 phyllum. In other individuals the carinae appear so early as 

 entirely to obliterate the Cyathophyllum stage. It is clear 

 that here are two types originating from a common ancestor, 

 which had both a Cyathophyllum and a Heliophyllum stage, 

 the latter being a late ephebic character. One retained the 

 Heliophyllum stage as an ephebic character, and is to be 

 regarded as a retarded type. The other became accelerated 

 until the carinae stage became established as a nepionic char- 

 acter. It is apparent that here are two distinct types, each 



* "Hamilton Group of Thedford, Ontario," Shinier and Grabau ; Bull. 

 Geol. Soc. Amer., vol. xiii, pp. 167-168. 



