The Scott is J i Naturalist. 161 



parts of ancient living things. We shall be unable to do more than 

 guess at their habits and mode of life, again a misfortune ; but 

 this is no leason, it seems to me, why we should omit to describe 

 those parts which are left, in a general account of the plant and 

 animal worlds. 



On the other hand the subject of Etiology can be discussed 

 with every prospect of attaining the most interesting and important 

 results, as instanced in the Etiology of the Herbivora and of the 

 birds as worked out by Huxley and Marsh respectively. 



It is in the subject of Distribution, however, that the importance 

 of uniting Palaeontology and Biology is especially apparent. 



Distribution may be either Geological or Geographical, that is 

 to say it treats of the distribution of plants and of animals in 

 time or in space. The true relationship of these two phases of the 

 Distribution question does not seem to me to be sufficiently well 

 brought out in textbooks. We are apt to think of Distribution in 

 time as merely the order in which different grades of organisms 

 have appeared on the globe, and Distribution in space as a census 

 of the inhabitants of the different sections into which the surface 

 of the land has been divided. But distribution in time is some- 

 thing more than that, and the two aspects of the subject, though 

 apparently as disconnected as depth and breadth, are in reality 

 inseparable. 



History has been defined as " statistics in movement ; " statis- 

 tics as "history in repose." And in geological history as in civil 

 history this is true. Geographical distribution is a statistic. Each 

 period in the history of the earth has its own geographical 

 distribution of organisms. These are all statistics, and since there 

 is an indefinite number of separate " horizons " in the succession 

 of rocks we may say there is an indefinite number also of statis- 

 tics of Geographical Distribution. The sum of them piled on 

 each other forms Distribution in time or Geological History. 

 Geographical Distribution as we understand it therefore is simply 

 the last census of living things, the first preserved otherwise than 

 by the rocks themselves. Here, again, imperfections in the 

 geological record, destruction of entire series of strata, and ignor- 

 ance of the geological history of the crust of the earth save in 

 certain parts, prevent us from attaining as perfect a knowledge of 

 past Geographical Distribution as we should wish, but nevertheless 

 the main features of Geographical Distribution in the Tertiary 



L 



