175 



scant numbers and of small size in Lake, Porter and Laporte counties. 

 This typical northern species may also be regarded its one that has 

 retained a foothold iu isolated localities after the recession of the shores 

 of Lake Michigan and the disappearance of the bordering swamps. The 

 reported occurrence of the species in Tippecanoe County (Golden) "in 

 sparing numbers along the Wabash River" demands further study. The 

 well-known difficulty of discriminating the species of Betula, due to sea- 

 sonal and age changes in appearance and habits, suggests that a closer 

 study may prove the reference an error. 



The Paper or Canoe Birch {Betula papyrifera Marshall) is found in 

 Lake and Porter counties, always being reported as rare and of small size. 

 This is another species definitely northern in its mass distribution, the 

 Indiana stations standing as its extreme southern limit. It is probably 

 another of the species which entered our area in the time of flooded rivers 

 and lakes of the Champlain but one which has been able to maintain a 

 precarious foothold up to the present time. Its early disappearance from 

 the tree flora seems inevitable. 



The Tag or Speckled Alder {Alnus incana (Linnaeus) Muenchhausen) 

 is found in Lake and Porter counties between dunes near to the lakes. 

 This is the common alder found in swamps and on the borders of streams 

 further north. It has been able to maintain itself in our area in greater 

 numbers and with less reduction of size than any other one of these 

 extreme northern species. 



A. dwarf shrubby form of Hackberry (Celtis pumila (Muhlenberg) 

 Pursh) is included. In both Gray and Sargent the form is regarded as a 

 variety of C. occidentalis Linnaeus. Its general range is in the South 

 Atlantic States ranging westward to Missouri, Colorado, Utah and Nevada. 

 It has been reported only from Lake County near the Calumet River at 

 Millers. Its occurrence in such a restricted locality is rather puzzling and 

 as yet no satisfactory explanation has been reached. The form in the 

 greater part of its range occurs on rocky banks of streams — a condition not 

 even approximated at its Indiana station. The temptation to regard it as 

 an ecologic variant of the very common Celtis occidentalis is almost irre- 

 sistible. 



Of the six species just discussed, five are definitely northern in their 

 mass distribution. Their presence as members of our flora is very evi- 

 dently referrable to the continuity of waterways existing during the 



