METHODS 

 Study Areas 



In 1928, 44 phenological observation stations were established throughout Montana and 

 northern Idaho, expanding to 46 in 1929. These included stations on 18 National Forests, at 

 Savenac Nursery in western Montana, Priest River Experimental Forest in northern Idaho, and 

 Yellowstone and Glacier National Parks. Geographically, they ranged from the Custer National 

 Forest in eastern Montana to the Kaniksu National Forest in northwest Idaho, an east-west 

 distance of about 700 air miles (1 100 km). Elevationally, they averaged about 3,200 ft 

 [975 m) and ranged from 2,200 to 4,150 ft (670-1 265 m) west of the Continental Divide. East 

 of the Divide the average was about 5,300 ft (1 615 m) with a range of 3,200 to 6,750 ft (675- 

 2 060 m) . By 1930, 35 of these stations were still reporting. According to Weidman, "Due to 

 the need for relief of ranger work loads, observations were abandoned on 12 forests between 

 1932 and 1934." The number of stations was reduced again in 1935, but a concentrated effort 

 remained to maintain some vegetative stratification in the white pine type (figs. 1, 2), the 

 ponderosa pine type (fig. 3), the larch-fir type (fig. 4), and the lodgepole pine type (fig. 5). 

 By the last year of this study, 1937, observations were still being conducted on the Bitter- 

 root, Clearwater, Gallatin, Flathead, and St. Joe Forests, and at Savenac Nursery, Priest River 

 Experimental Forest, Deception Creek Experimental Forest, and Yellowstone National Park. 



As shown in figure 6, the phenological stations were not only widely dispersed geograph- 

 cally, but also fairly well distributed. 



Data Collections 



As many as 16 different phenological events were recorded at the outset of the study in 

 1928. During the next 10 years this was gradually reduced to the more meaningful and measur- 

 able events--10 events in the conifers, 9 in the hardwood trees and shrubs, and 11 in the herbs. 

 Included were events such as bud burst, pollen shedding, seed fall, leaf fall, and others.^ 



At each phenological station, conifers, hardwood trees and shrubs, and herbaceous plants 

 common to the locality were measured phenologically. In 1928, sampling included 12 conifers, 

 16 hardwood trees and shrubs, and 14 herbaceous species. As discussed later, this list was 

 expanded during the following 10 years. Some stations had few species, and some had many, but 

 none had all of the above combination of species represented. 



Phenological stations were established near ranger or other forest headquarters occupied 

 during the growing season, near a weather station, and in a spot "representative" of the forest 

 type of the area. One or more persons at each study area were assigned to record the pheno- 

 logical events weekly by species. A study plan by R. H. Weidman in 1928 detailed the methods 

 for all phenological stations, serving as a common guide for all observers. Observers selected 

 were to be "men who have enough practical knowledge of botany to make reliable records, and 

 who at the same time are personally interested in making observations of this sort." In 

 reality, weekly measurements were simply not possible on all areas, and in some cases a whole 

 season was missed because of insufficient staffing or higher priority work schedules. 



Analysis Methods 



In 1933, Ellison prepared an internal report of the results of 5 years of data from six 

 phenology stations in forests of the white pine type. He searched for relationships to 

 illustrate "fundamental principles as to the relation of vegetative events and the causal 

 factors of air and soil temperature." He felt unsuccessful in this effort. As a result, 

 nothing was published. Subsequent internal discussions and reports by Watts, Koch, Haig, and 

 Weidman pointed out that the "practical usefulness of simply knowing earliest, latest, and 

 average dates of beginning and ending of vegetative activities" was the most important infor- 

 mation from the study. Again, nothing was formally published and interest in phenological 

 studies apparently waned. 



See the appendix for definitions of the phenological events used in this study. 



2 



