September 17, 1915] 



SCIENCE 



367 



The belt number can conveniently be distin- 

 guislied from its division numbers by a dot 

 placed between tbem. Thus the state of 

 Georgia would be included for the most part 

 in N 6.10 with the extreme eastern area in 

 N 6.9. The island of Tasmania would be 

 comprised in S 7.17 and S 8.17. 



Subdivided in the above manner, the United 

 States would be comprised in belts N 5 to N 9 

 and would be contained in 29 divisions. 



The above-defined areas, though large, are 

 sufficient to indicate in a general way the dis- 

 tribution of plants and animals. But where 

 greater exactitude is required, as, for example, 

 where it is desired to indicate the most south- 

 erly point reached by a typical northern spe- 

 cies or vice versa they are rather vague. Ac- 

 cordingly each " merosphere " may be again 

 divided into smaller areas, each consisting of 

 one degree of latitude and one degree of longi- 

 tude. As will be seen from the annexed figure 

 the east and west sub-belts are numbered from 

 1 to 6 and the nine strips running north and 

 south are marked from A to I. By the com- 



bination of a number and letter the position 

 of each of the 54 subdivisions can be indi- 

 cated, thus N6.10.5E wiU include the north- 

 west corner of Georgia. One advantage pos- 

 sessed by the system of numbered areas herein 

 described is that it can be used alone or in 

 conjunction with the name of the state or 

 province where fuller information is desired. 

 Thus Georgia (]Sr6.9.3I) will indicate the area 

 around Savannah. Altogether this state con- 

 tains twenty-four of these smaller subdivisions. 



If some such scheme of uniform subdivisions 

 were adopted in the case of a " flora " which 

 deals with a single state or country the sepa- 

 rate units of adjacent states or countries could 

 readily be grouped into the larger areas men- 

 tioned. 



It must, of course, be borne in mind that 

 some such scheme as the above is intended to 

 indicate only the geographical range of spe- 

 cies. The division of the earth's surface into 

 hiological provinces, each with its characteris- 

 tic assemblage of plants and animals, is a dif- 

 ferent matter. Before we can divide the earth 

 satisfactorily into biological areas we must 

 know first the actual limits of distribution of 

 each species as well as something of the clima- 

 tology in the widest sense of the proposed 

 areas. At present we have not amassed suffi- 

 cient data to enable us to make these wide 

 generalizations, though several persons have 

 made the attempt with varying success. 



The need for some such division of the 

 earth's surface, as I have attempted to out- 

 line above, is not so evident in the United 

 States as it is further north. Such terms as 

 Labrador, Alaska and Ontario, referring as 

 they do to very large areas, are too indefinite 

 to indicate with exactness the distribution of 

 the fauna and flora. Labrador formerly had 

 a difl^erent meaning from what it has at pres- 

 ent, the greater part of the territory formerly 

 called by that name being now part of the 

 province of Quebec, and even yet the boundary 

 line is not properly delimited. 



The distribution of a certain species is given 

 in the " North American Mora " as " Nova 

 Scotia to Georgia, Tennessee and Michigan." 

 Presumably it occurs in Ontario, as a line 

 drawn from Nova Scotia to Michigan will pass 

 through the province of Ontario. But it is 

 not by any means certain whether the species 

 occurs in the southern parts of New Bruns- 

 wick or Quebec. According to the method 

 outlined above the distribution would be N6.9- 

 10, N7.8-10, N8.8-10. 



Another species is mentioned as extending 

 " from Newfoundland to Florida, Alabama 

 and the Mackenzie." As this river has a 

 course of over 1,000 miles in length it will be 



