318 REPORT OF NATIONAL MUSEUM, 1892. 



during' that age they were comparatively unimportant. Although -fishes 

 existed earlier than other vertebrates, their remains afford little material 

 for comparison with those of invertebrates in earlier rocks than those 

 of Devonian age. Therefore the identification of all those earlier forma- 

 tions, comprising a large part of the whole geological series, must be 

 made by means of invertebrate remains alone. 



When considering the relative value of fossil remains of animals and 

 plants in the identification of formations, only terrestrial plants* are 

 worthy of notice, because of the low grade of aquatic plants proper, and 

 because our definite knowledge of terrestrial plants extends no farther 

 back in geological time than the Devonian. Furthermore, their remains 

 being very rare in rocks of that age, a satisfactory comparison of plant 

 remains with those of animals is necessarily restricted to formations of 

 the Carboniferous and later ages. 



Again, the disparity of value between the remains of aquatic and 

 terrestrial faunas, which has already been mentioned, is not only 

 equally great between those aquatic faunas and laud floras, but for 

 reasons stated in Essay i plant remains in all cases much less com- 

 pletely represent the floras from which they were derived than do the 

 remains of aquatic animals the faunas in which they originated. 



From the foregoing statements it is seen that, although fossils in all 

 cases constitute not only much the most, but usually the only, trust- 

 worthy criteria for such identification of format ions as is indispensable 

 in the study of structural geology, the various kinds differ materially 

 as to their relative value. This value, however, has no necessary rela- 

 tion to that which they may possess as indicators of geological time or 

 of the correlation of the strata containing them with those of other 

 parts of the world. The two values are distinct, although one kind of 

 fossil remains may often possess both. 



Other criteria of identification than those of fossil remains were 

 briefly referred to in a preceding paragraph as being chiefly homogeny 

 and lithological similarity, both of which are valuable aids when pale- 

 ontological criteria are deficient, and both may often be relied upon in 

 cases of the absence of those criteria. Except in the more or less con- 

 stant use of lithological characteristics, that ought always to be made 

 in connection with the biological identification of formations, those 

 characteristics are at best of limited applicability as criteria, and they 

 are available only in either direct or indirect relation to biological cri- 

 teria. Their indirect availability is that which has just been referred 

 to. Their direct availability is mainly in connection with what in a 

 preceding paragraph has been designated as relative identification, be- 

 cause it can be made only with reference to some one or more of already 

 known biological horizons. 



Sometimes the relative strati graphical position of a single unfossilifer- 

 ous sedimentary formation may be determined by means of its relation to 



Palustral plants are included in this designation. 



