348 EDWARD W. BERRY 



of central and southern Mexico and the highlands of Guatemala. 

 While its range is extensive and its habitat varied it reaches the 

 largest size and commercial possibilities in the rich bottom lands 

 of mixed hardwoods in the maritime districts or coastal plain of 

 our southern states from the valley of the Great Pedee in South 

 Carolina to the valley of the Trinity River in Texas, and north- 

 ward along the bottoms of the Mississippi, Arkansas, Tennessee 

 and Ohio. In the more northern part of its range it inhabits 

 swamp borders and low wet swales. 



There are at least three additional existing species of Liquidam- 

 bar — one Liquidambar macrophylla Oersted, found in the moun- 

 tains of Central America, while the other two are asiatic. Li- 

 quidambar formosana Hance is found on the island of Formosa 

 and in southern China, and the other separated from it by the 

 whole breadth of the Asiatic continent, is found in a limited area 

 in the mountains of southwestern Asia Minor. The last was 

 named Liquidambar orientalis by Miller and is the source of the 

 liquid storax of commerce. 



This disconnected distribution of the existing species of Liq- 

 uidambar, which can be better appreciated by a glance at the 

 accompanying map, figure 1, is a sure indication of an ancient 

 lineage and a former occupation of the intervening areas where it 

 is now extinct. If the sweet gum stood alone in having such a 

 remarkable range its interest would seem much greater, but since 

 the days of Asa Gray's American Association address we have 

 become accustomed to many similar ties across the departed 

 ages that formerly connected and now explain the near kin found 

 in Asia and North America exemplified also by the magnolia, 

 sassafras, coffee-bean and tulip tree. 



Turning to the fossil record we find that about twenty extinct 

 species of Liquidambar have been described. The oldest of 

 these, Liquidambar integrifolius, described by Lesquereux from 

 the Upper Cretaceous Dakota sandstone of Kansas and subse- 

 quently identified from Canada, Texas and South America, can- 

 not be looked upon as the Abraham of the race of gums for un- 

 fortunately for our story its coriaceous and entire-margined 

 leaves are not those of a Liquidambar but probably represent a 



