490 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



Nebraska and part of South Dakota. In my opinion, these parts 

 do not belong to the Upper Sonoran Zone. In the Bulletin men- 

 tioned, I can not find any statement giving the reason why this 

 region is included in the Upper Sonoran Zone, and why the north- 

 ern part, viz. northeastern Montana, eastern Alberta, Saskatche- 

 wan, and western North Dakota, is excluded. Perhaps the zone 

 is based wholly on animal distribution, for the flora does not 

 warrant such a di\-ision. Maybe the occurrence of austral types 

 in this region has lead to the error. But then the question arises, 

 were these austral types Sonoran or were they from the eastern 

 Upper Austral Zone. The true prairie region belongs to that 

 zone and it extends along the Missouri \'alley through South 

 Dakota into Montana and along the North Platte and Cheyenne 

 River valleys into W yoming. 



The flora of the Great Plains from the Arkansas Divide in Col- 

 orado to the divide between the Saskatchewan and Athabasca 

 rivers in the north, and from about the ioo° meridian to the foot- 

 hills of the Rockies, is practically the same. The characteristic 

 grasses, Koeleria nitida (vernal facies), and the two gramas, 

 Bouteloiia oligostachya, B. hirsuta (autumnal facies) on the plains 

 proper, Andropogon scoparius and Bouteloua curtipendula on the 

 hills, and Agropyron Smithii, A. molle, etc., in alkaline soil, are 

 found from southern Colorado to Saskatchewan, and the Buffalo 

 grass, Bulhilis dactyloides, is scattered over the region, except in 

 the northern portion. The common trees and shrubs of the hill- 

 sides, Padus melanocarpa, Rhus trilobata, Toxicodendron Rydbergii, 

 Grossidaria odorata, and Symphoricarpos occidentalism extend north 

 to northern Montana or Alberta. It is true that some of these, 

 as for instance Bouteloua oligostachya, B. ciirtipendnla, and B. 

 hirsuta are also found in the Sonoran Region, luit they are the only 

 ones among their numerous relatives which are found on the 

 plains north of the Arkansas Divide. Of the cactuses, Opiintia 

 polyacantha and Cactus viviparus range from Colorado to Sas- 

 katchewan and Opuntia fragilis and Cactus Diissoun'oisis to 

 northern Montana. 



South of the Arkansas Di\ide are found many plants either not 

 found IK nth (hereof or not extending far north thereof. Among 

 the grasses ma>- be mentioned Amphilophis Torrcyanus, Bouteloua 

 polystachya, B. criopoda, B. prostrata, Scleropogon brcvifolius, and 



