GEOLOGY — WILLIS. 207 



it necessary to state alternative hypotheses in cases where the conclu- 

 sion can best be left to a more advanced state of knowledge ; but it 

 is to be expected that we may gain as we put the alternatives. 



The other condition of indeterminate questions is incompleteness of 

 facts. It is partly incompleteness of the geologic record, from which 

 many sections are missing ; it is also partly the state of geologic ex- 

 ploration, which still fails to cover, in an adequate degree, large areas 

 of several continents. We can not restore the missing parts of the 

 record, yet a fuller assemblage of the existing parts should enable us 

 better to understand what the}' might signify. We can not attempt 

 to survey the world, but by studying the available data we shall arrive 

 at a better knowledge of the critical districts where surveys will reap 

 the largest har\'est, and some part of these surveys we may make. 



METHODS OF RESEARCH. 



The methods of the proposed research comprise (a) assembling 

 data of stratigraph3% paleontology, and orogeny ; (d) interpretation ; 

 (r) correlation ; (d) delineation of results on maps and discussion of 

 mooted points in manuscript. 



Assc77iblmg. — To assemble the data requires extensive reading, 

 which should be planned, first, with reference to a definite province, 

 and, second, according to geologic epochs. A record should be kept 

 in classified notes and on index maps, and these will serve as a basis 

 of compilation on provisional and final maps. 



Assume that we are considering the Appalachian province of North 

 America, and are beginning with New York State, the best-known 

 district. The geologic map gives the distribution of formations of 

 many kinds and many epochs. Each formation exhibits characters 

 (constitution, texture, color, bedding, surface markings, and con- 

 tained fossils) which are peculiar to the conditions of origin of 

 material, to the conditions of transportation, and to the conditions of 

 deposition. The interpretation of the.se will be considered in the 

 next paragraph. We are now concerned with assembling the facts. 

 The geologic map is a composite picture. To analyze it we select 

 formations which mark successive geographic conditions and map 

 each formation separateh', noting not only its distribution, but also 

 variations in its characters. The data must be gathered not only 

 from the original geologic map, but also from the literature. There 

 will result a body of material (maps and notes) which should be 

 kept classified for convenient reference and comparison, as the study 

 extends to other districts of the province. In so large a work as is 

 here proposed, to extend over years, it is important to arrange the 



