1875.] 4l'±J [Stevenson. 



a long period of time. Tlie lagoon theory is quite ingenious, but unfor- 

 tunately cannot accommodate itself to the facts. 



Some species of invertebrates showed remarkable tenacity of life. Thus 

 Stropliomena rliomboidalis reaches from the Lower Silurian quite to the 

 base of the Lower Carboniferous. Atrypa reticularis existed from near 

 the beginning of the Upper Silurian to near the close of the Devonian. 

 In each group they show marked peculiarities which almost suffice to 

 mark the horizon from which the specimens were obtained. But no pa- 

 laeontologist would be reckless enough to determine a horizon with these 

 shells as his only data. While we find instances of this kind passing up- 

 ward, we have never found characteristic Carboniferous species in lower 

 formations. But if we should, we must yield to the superior evidence. 

 Spirifer cameratus associated with a strongly marked Devonian fauna in 

 rocks occupying the Devonian position, would be a worthless witness. 

 So, if the thing were possible, should we find Ammonites at a Silurian 

 horizon, we would reject the testimony in favor of Mesozoic and accept 

 the stronger testimony for Silurian. Even invertebrate life must yield to 

 stratigraphy, if the two contradict. 



Vertebrate life is too imperfectly preserved to be ordinarily of much' 

 service alone. The succession is not fully given. Yet it may be service- 

 able. If certain reptilian forms are found constantly associated with a 

 certain invertebrate fauna, as, for example, certain forms in the Creta- 

 ceous, we may accept those as evidence where other evidence is wanting, 

 for their horizon has been definitely fixed. This, however,- applies only 

 to marine forms. To terrestrial forms, the same objection applies as to 

 plants. In every case, however, the horizon must be fixed for a conti- 

 nent, not for the world, since the conditions affecting such life may have 

 been different in America from what they were in Europe. 



Vegetable life shows no such history as to entitle it to much considera^ 

 tion. So patent is this fact that little use has been made of vegetable re- 

 mains in determining the succession of rocks. Fucoids are worthless ex- 

 cept in limited areas, since their organization is so low as to enable them 

 to withstand changes which would be fatal to higher organisms. Land" 

 plants are unsatisfactory, because they are preserved in disconnected 

 fragments, and because the areas on which they grew were so widely 

 separated and formed so small a portion of the earth's crust. Let us look 

 at the succession as we find it. 



Until a very short time ago the existence of land plants during the Si- 

 lurian and early Devonian of America was denied, and some told us why 

 no such fiora could exist. Two water-worn logs of coniferous wood, 

 found in the Corniferous Limestone, changed our belief, but gave us little 

 information. Respecting the fiora of the Middle and Upper Devonian 

 and of the Lower Carboniferous, we have but limited knowledge, and the 

 localities yielding specimers are few indeed. Of the Coal Measure vege- 

 tation we know quite well that portion which grew in the swamps, but 

 of the upland flora we have only fragmentary information, in the shape of 



