Components of Adaptive 

 Variation in Pinus contorta 

 from the Inland Northwest 



G. E. Rehfeldt 



INTRODUCTION 



Adaptive differentiation is discernible either directly as 

 differential fitness in contrasting environments or in- 

 directly as genetic responses that parallel environmental 

 gradients. In Pinus contorta Dougl., for example, four 

 geographic races are distinguished both morphologically 

 (Critchfield 1957) and enzymatically (Wheeler and Guries 

 1982). Critchfield (1980) reviewed numerous studies that 

 provide direct and indirect evidence of differential adapta- 

 tion of races to climates as diverse as those of the north- 

 ern Pacific Coast, the Sierra Nevada, the interior Rocky 

 Mountains, and the Canadian subarctic. 



In P. contorta spp. latifolia, the subject of this paper, 

 genetic variation among populations is pronounced. Cana- 

 dian populations, native to a region of dissected plateaus, 

 are arranged along gentle clines that follow climatic 

 gradients across 18° of latitude from the Yukon to 

 southern British Columbia (Hagner 1970; Lindgren and 

 others 1976; Ying and others 1985). In the mountains of 

 Idaho (Rehfeldt 1983a), Utah (Rehfeldt 1985a), and Oregon 

 (Stoneman 1985), population differentiation occurs along 

 steep elevational clines that parallel the environmental 

 changes associated with altitude. Whether geographic or 

 elevational clines predominate, populations from mild en- 

 vironments express a high innate growth potential and low 

 cold hardiness, while those from severe environments 

 display a low growth potential and high hardiness. 



Adaptive clines result from environmental selection of 

 phenotypes that, for long-lived trees, develop in environ- 

 ments of extreme temporal heterogeneity. Adaptedness, 

 therefore, has many component traits. A consideration of 

 the interrelationships among components inspired Lande 

 (1982) to argue that negative genetic correlations are 

 often obscured phenotypically but commonly set the limits 

 on microevolution. In trees, for example, negative genetic 

 correlations relate growth potential and cold hardiness 

 within families of both Pseudotsuga menziesii (Mirb.) 

 Franco and P. contorta (Rehfeldt 1984). This means that 

 an assessment of adaptive differentiation requires an 

 understanding of component traits and their interrelations. 

 Lande (1982) stressed that this understanding is par- 

 ticularly necessary for fields such as forestry where the 

 traits of agronomic importance— growth and yield— are 

 also major components of fitness. ' , 



The present study of Pinus contorta assesses genetic 

 variability among populations, relates genetic variability to 

 adaptive differentiation, and presents adaptive landscapes 

 for the Inland Northwest. 



DISTRIBUTION, ECOLOGY, AND 

 DEMOGRAPHY 



Adaptive variation must be interpreted according to the 

 spatial environmental heterogeneity within the region of 

 study (fig. 1). The chmate of the Inland Northwest varies 

 from the continental in the east to that with a coastal 

 component in the west (Daubenmire and Daubenmire 

 1968; Pfister and others 1977). This general trend is 

 reflected not only in gradients of temperature and precipi- 

 tation (fig. 2), but also in the composition and distribution 

 of plant communities (Daubenmire and Daubenmire 1968; 

 Pfister and others 1977;. Steele and others 1981). Superim- 

 posed on general climatic gradients are the topographic 

 microclimates associated with mountainous terrain. Ex- 

 treme environmental heterogeneity thus develops from a 

 climatic transition that occurs across a series of rugged 

 mountain ranges. 



In the Rocky Mountains, P. contorta displays such a 

 broad ecological distribution that it is capable of growing 

 in almost any forest environment (Pfister and Daubenmire 

 1973). As an early successional species, the pine is com- 

 mon in a variety of plant communities that include associa- 

 tions dominated at maturity by Tsuga heterophylla (Raf.) 

 Sarg. on mesic sites, Pseudotsuga menziesii (Mirb.) Franco 

 on dry sites, and Abies lasiocarpa (Hook.) Nutt. on cold 

 sites (Daubenmire and Daubenmire 1968; Pfister and 

 others 1977; Steele and others 1981). The species forms 

 large continuous populations on subalpine sites at eleva- 

 tions up to 3,000 m and in frost pockets on valley floors as 

 low as 600 m. ' ' - - 



Natural populations of P. contorta tend to be established 

 in cycles (Lotan and others 1985). Greatly simplified, these 

 cycles involve (1) wildfire, (2) profuse even-aged reproduc- 

 tion of as much as 500,000 seedlings/ha from either open 

 or serotinous cones (Tackle 1959), (3) intense natural thin- 

 ning, which can leave as few as 1,000 trees/ha by age 80 

 (Tackle 1959; Benson 1982; Vyse and Navratil 1985), and 

 (4) epidemics of the mountain pine beetle {Dendroctonus 

 ponderosae Hopkins) (Amman and others 1973; Shrimpton 

 and Thomson 1983), which supplement the competitive 

 mortality to provide the fuel for (5) wildfire. 



These demographic cycles have pronounced effects on 

 the genetics of populations. First, adaptive traits will in- 

 clude all traits that either directly or indirectly influence 

 the expression of growth potential and thereby determine 

 which trees are living when fires occur. And second, pop- 

 ulations are frequently established on the same sites on 



1 



